datacontract-edit

name

datacontract-edit

description

Edit an output-port ODCS file under models/output_ports/v<N>/, run the contract test against the live data, and classify any failures as breaking or non-breaking changes — with suggested fixes. Only edits output-port contracts (the spec this data product commits to); input-port contracts under models/input_ports/ are upstream's responsibility and refreshed by dataproduct-implement. Trigger when the user asks to "add/remove/change a column in the data contract", "update the data contract", or "test contract changes".

Edit a data contract and test the impact

Change an output-port models/output_ports/v<N>/*.odcs.yaml file, run the contract test, and tell the user whether the change breaks consumers.

This skill operates only on output-port contracts — the spec this data product commits to. Input-port contracts under models/input_ports/ are cached snapshots of upstream's spec and are not editable here; if you want to refresh one (because upstream changed it), run dataproduct-implement instead.

How to run this skill

${PLUGIN_ROOT} below refers to the root of this plugin — the directory that contains skills/. On Claude Code it is set automatically as ${CLAUDE_PLUGIN_ROOT} — use that. On any other agent (Codex, Copilot CLI, etc.) it is unset; resolve it as ../.. relative to this SKILL.md file's directory (i.e. the grandparent of skills/<this-skill>/).

Plan announcement (before Step 0)

Before running Step 0, print this plan to the user verbatim:

Running datacontract-edit. I'll:

1. Locate the output-port contract file under models/output_ports/v<N>/ that matches your request.
2. Apply the edit in place and show you a unified diff.
3. Run datacontract test against the live server to check the change.
4. Classify each failure as breaking-schema, breaking-quality, additive, or unrelated.
5. Report and suggest concrete fixes (no version bump, no v2 directory, no dbt model changes).

Then proceed.

Step 0 — Locate the contract

• Search only models/output_ports/**/*.odcs.yaml — never models/input_ports/. If the user names an input-port contract, stop and explain it can't be edited here (refresh via dataproduct-implement instead).
• If the user named a contract file or column, find the matching file under models/output_ports/.
• If multiple contracts exist and it's ambiguous, list them and ask which one.
• Read the file and remember the current models block as BEFORE.

Step 1 — Apply the edit

Edit the ODCS YAML in place using the user's instruction. Keep the change minimal — do not reformat unrelated fields.

Common edits and the right shape:

User says	What to change
"add column X"	Append a field under the relevant model with at least type; set required: false by default unless the user says it's required
"remove column X"	Delete the field; this is breaking — flag in Step 4
"rename X to Y"	Rename the field; this is breaking — flag
"make X required"	Add required: true; breaking if existing rows can be null
"change X type from int to string"	Update type; breaking unless the new type is a strict superset (e.g. int → bigint)
"add a unique/not_null/enum check"	Add to the field's quality rules; breaking iff existing data violates the new rule — only Step 2 can tell

After editing, remember the new models block as AFTER and show the user a unified diff before continuing.

Step 2 — Run the contract test

Run the test with the datacontract CLI against the local contract file:

datacontract test models/output_ports/v<N>/<file>.odcs.yaml --server <server> --logs

• If the contract has more than one server, ask which one (typically production). Default to all only if the user explicitly asks.
• Use --logs so failure detail is in the output you read; otherwise the CLI only prints a summary.
• If the user wants the report persisted, add --output ./test-results/junit.xml --output-format junit.
• Capture stdout + exit code as TEST_RESULT. Non-zero exit means at least one rule failed; the log section names the failing field/rule.

Pre-reqs the CLI needs (verify before running, fail fast with a clear message if missing):

• uv run --quiet datacontract --version succeeds from the project root. If it fails, run uv sync and retry. Invoke as uv run datacontract test … for every CLI invocation in this skill.
• The chosen server's credentials available as env vars per the ODCS server block (e.g. DATACONTRACT_SNOWFLAKE_USERNAME / ..._PASSWORD, or DATACONTRACT_DATABRICKS_TOKEN). Tell the user which env vars are missing — do not try to source them yourself.

Do not use the platform's server-side contract-test endpoint from this skill. The local datacontract CLI runs against the edited file and gives line-level failure detail; testing the published version on the server would test the previous contract, which defeats the point of testing the edit.

Step 3 — Classify the outcome

Group every failure into one of these buckets:

Bucket	Examples	Severity
Breaking — schema	column removed, type narrowed, column renamed	High — bump major version, deprecate old port
Breaking — quality	new not_null/unique/enum rule violated by existing data	High — clean data first, then re-test
Non-breaking — additive	new optional column, widened type, loosened rule	Low — minor version bump
Test failure unrelated to the edit	flaky source, infra error, unchanged rule failing	Investigate separately

For each failure, name the exact field/rule and which bucket it falls into. Don't lump them together.

Step 4 — Report and suggest fixes

End with this two-part recap. The Status column uses the shared enum (created, updated, already present, deferred, skipped), and below it a classification table covers any test failures.

Part 1 — outcome table.

Artifact	Status	Details
Contract file	updated	models/output_ports/v<N>/<file>.odcs.yaml — show the unified diff inline
Contract test	…	pass, fail (<N> failures), or not run (missing creds) — name the server
Breaking — schema	…	count of failures in this bucket, or "—"
Breaking — quality	…	count of failures in this bucket, or "—"
Non-breaking — additive	…	count of changes in this bucket, or "—"
Test failures unrelated to the edit	…	count, or "—"
Recommended version bump	…	patch / minor / major based on the edit

For the four "bucket" rows, leave Status = — and put the failure count + a one-line bucket description in Details.

Part 2 — next steps. Per failure, give a concrete fix suggestion:

• Schema breaking: bump to a new output port version (models/output_ports/v2/), keep v1 alive, add a deprecation note in <id>.odps.yaml.
• Quality breaking: SQL snippet to find the offending rows so the user can decide whether to clean, accept, or relax the rule.
• Additive: bump the contract's version minor, no consumer impact.
• If no failures, recommend the version bump (patch for cosmetic, minor for additive, major for breaking even when the test happened to pass).

Do not auto-bump the contract version, do not create v2/ directories, and do not modify dbt models. Surface the recommendation; let the user decide.

Constraints

• Always run the test after every edit. A passing edit looks the same as a breaking edit until you run it.
• Don't edit the dbt models in this skill. Changing the contract is a separate decision from changing the implementation. If the user wants both, run dataproduct-implement after this skill.
• Don't fetch a remote version of the contract — operate on the local file. If the user wants to pull the published version, ask them to do that explicitly first (it can be a one-line entropy-data datacontracts get <id> -o yaml redirected to the file).
• Idempotent: re-running with the same edit should be a no-op (same diff = empty, same test = same result).